摘要目的：探讨MRI定量增强参数与乳腺癌预后因子及分子分型的相关性。方法回顾性分析经手术病理证实为乳腺癌，行乳腺MRI定量动态增强扫描，且术后经免疫组织化学检测的87例患者（91个乳腺癌病灶），其中80例（84个病灶）获得分子分型结果。所有患者均行乳腺MRI平扫及增强扫描，并测量容量转移常数（Ktrans）、速率常数（kep）和血管外细胞外间隙容积比（Ve）。记录免疫组织化学结果，观察雌激素受体（ER）、孕激素受体（PR）和人类表皮生长因子受体?2（HER?2）表达情况，并对患者进行分子分型，同时观察预后因子Ki?67的表达情况。87例患者的91个乳腺癌病灶中，ER阳性病灶68个，阴性病灶18个；PR阳性病灶60个，阴性病灶26个；HER?2阳性病灶30个，阴性病灶58个（其中部分患者未同时获得ER、PR及HER?2受体表达结果）。确定分子分型的80例（84个病灶）患者中，Luminal A型19个，Luminal B型49个，HER?2阳性型7个，三阴性乳腺癌（TNBC）9个。各受体在不同表达状态下的MRI定量增强参数采用Mann?Whitney U检验比较，定量参数和Ki?67阳性表达率的相关性采用Spearman方法评价，不同分子分型乳腺癌患者的定量参数采用Kruskal?Wallis检验比较。结果 ER、PR阴性患者的Ktrans及kep值高于ER、PR阳性患者，差异有统计学意义（P<0.05），HER?2阳性和阴性患者的参数值差异均无统计学意义（P>0.05）。Ki?67阳性表达率为5%～90%（中位数为33%），Ktrans、kep及Ve值的中位数及四分位间距分别为1.33/min（0.88/min）、2.63/min（2.34/min）、0.51（0.22）。Ki?67阳性表达率与Ktrans、kep值无明显相关性（r值分别为0.24、0.22，P值分别为0.03、0.04），和Ve间无相关性（r=0.00，P=0.97）。不同分子分型病灶间Ktrans值、kep值差异有统计学意义（P<0.05）， Luminal A、Luminal B型的Ktrans值低于HER?2过表达型及TNBC型，Luminal A型及Luminal B型的kep值小于TNBC型，而Ve值则均高于TNBC型，Luminal B型的kep值低于HER?2过表达型，差异均有统计学意义（P<0.05）。TNBC的kep值在各型中最高（3.99/min），而Ve值则最小（0.41）。结论通过MRI定量增强参数可在一定程度上提示乳腺癌不同亚型，和预后相关因子不相关。

abstractsObjective To investigate whether correlation exists between quantitative perfusion parameters obtained from dynamic contrast?enhanced magnetic resonance imaging (DCE?MRI) and different prognostic factors or immunohistochemical subtypes of breast cancers. Methods A retrospective analysis of DCE?MRI was performed in 87 breast cancer patients with 91 pathology confirmed breast lesions. All the patients underwent the immunohistochemistry after the operation, 80 breast cancer patients with 84 breast lesions were divided into different subtypes based on the immunohistochemical profiles. All the patients underwent plain MRI and DCE?MRI. The volume transfer constant (Ktrans), rate constant (kep) and volume of EES per unit volume of tissue (Ve) value were calculated. Results of the immunohistochemistry and observation results of the the expression of estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor?2 (HER?2) were recorded. Subtypes were also categorized. The degree of the expression of Ki?67 was observed simultaneously. Among the 87 breast cancer patients (with 91 breast&nbsp;lesions), 68 were ER?positive,18 were ER?negative, 60 were PR?positive, 26 were PR?negative;30 were HER?2 positive and 58 were HER?2 negative. Eighty breast cancer patients (with 84 breast lesions) were divided into different subtypes:19 were Luminal A, 49 were Luminal B, 7 were HER?2 positive subtype and 9 were triple negative breast cancer (TNBC). The quantitative perfusion parameters with diffierent expression of the receptors were analyzed through Mann?Whitney U test, the correlation between the degree of the expression of Ki?67 and quantitative perfusion parameters was obtained with Spearman test. Furthermore, the distributions of the perfusion parameters between groups according to immunohistochemical subtype were compared using the Kruskal?Wallis test. Results Ktrans and kep value were higher in tumors with ER?negative and PR?negative than with ER?positive and PR?positive and difference was statistically different ( P<0.05). There was no statistically significance in the quantitative perfusion parameters between HER?2 positive and HER?2 negative (P>0.05). The degree of the expression of Ki?67 was 5%to 90%the mean value was 33%, the median and interquartile range of Ktrans, kep and Ve value was 1.33/min (0.88/min), 2.63/min (2.34/min) and 0.51 (0.22) respectively. There was no significant correlations between the degree of the expression of Ki?67 and Ktrans and kep value (r value was 0.24 and 0.22 respectively, P value was 0.03 and 0.04 respectively), and the degree of the expression of Ki?67 was not associated with Ve (r=0.00, P=0.97). Ktrans value was lower in Luminal A and Luminal B than that in HER?2 positive subtype and TNBC, kep value was lower in Luminal A and Luminal B than that in TNBC, and Ve value in Luminal A, Luminal B and HER?2 positive subtype were higher than that in TNBC, kep value was lower in Luminal B than that in HER?2 positive subtype, statistically significance was seen among them (P<0.05).Triple?negative breast cancer showed the highest kep value (3.99/min) and the lowest Ve value (0.41) than the other subtypes. Conclusion The different subtypes of breast cancer may be predicted to some extent with quantitative perfusion parameters, and there is no correlation between quantitative perfusion parameters and the prognostic factors.